The present invention relates to an Ethernet communication redundancy method and, more particularly, to an Ethernet communication redundancy method of making a communication system redundant between a plurality of nodes connected to each other through Ethernet communication lines to constitute a distributed control system.
In general, in a distributed control system (DCS) for controlling each node to be controlled by using a corresponding one of a plurality of nodes (terminals), a communication system for connecting the respective nodes is made redundant to increase the operation rate of the overall system, thereby efficiently controlling the system. When a communication scheme using a token, e.g., token passing, is used as a communication scheme between the respective nodes, since the cyclic period of the token can be calculated, disconnection in the communication system or the like can be detected by checking this cyclic cycle.
Assume that a communication scheme using such a token is used as the communication system of a distributed control system. In this case, a system and method which switch redundant systems for communication in response to disconnection or a communication failure in this communication system as a trigger are conceivable.
As personal computers become widespread and improve in performance, personal computers connected as nodes through Ethernet are now in increasing demand. The communication system of a distributed control system may therefore be made redundant by using Ethernet widely used in the field of personal computers.
In such a conventional redundancy method, however, when one Ethernet communication line is used for a communication system, since the communication line is shared by a plurality of nodes, the time taken for communication greatly varies due to contention control and the like. For this reason, this method cannot be used for a distributed control system which is required to exchange measurement data and control commands without any great delay.
For example, in Ethernet, CSMA/CD (Carrier Sense Multiple Access with Collision Detection) is used as a contention control method. According to this method, when one communication line is shared by a plurality of nodes, data is transmitted to the communication line after it is determined that the communication line is not in use. If, therefore, different nodes simultaneously start transmitting data, a collision occurs.
The probability of occurrence of collision increases as the number of nodes and the communication frequency increase. When a collision occurs, retry is performed (for example, 16 times=several hundred ms on the MAC layer). In this case, the time required for communication greatly varies. This method cannot therefore be used for process control that is required to exchange measurement data and control commands without any great delay.
In a redundancy scheme designed to switch communication lines, when a failure occurs in one of the communication lines, the failed communication line must be repaired while communication is performed exclusively through the normal communication line.